Turbotodd

Ruminations on tech, the digital media, and some golf thrown in for good measure.

Archive for April 22nd, 2011

Defending The Fortress

leave a comment »

I believe it was Napoleon who said “Never interrupt your enemy when he is making a mistake.”

Ron Markezich, the sales lead for enterprises and partners in the U.S. at Microsoft, apparently explained to AllThingsD earlier this week that for every dollar companies spend on Microsoft software, they have to spend $6 to get it running right.

So I went and read the interview to get the full context.

The specific question Markezich was asked centered around Microsoft’s disruption of its traditional business model that “has brought in billions upon billions of dollars is sold” and whether or not “this new model [cloud delivery] ultimately catch up with and supplant the old one?”

Here’s the latter part of his response:

…So every one of these customers, we see their total spend with Microsoft go up anywhere from 2 to 6 times what it was before. The other thing is that if you look at the total industry spend, most of it is on activities where there’s no value added. Every dollar you spend on software from Microsoft, you spend $6 trying to get it to do anything. What we’re trying to do is drive that six dollars to zero.

It was also Napoleon who explained that “the fate of a nation may sometimes depend upon the position of a fortress.”  Insert “business” for “nation” and the intent of the statement remains.

But after the last two days, some perhaps might ask if the cloud ain’t all it’s cracked up to be, either.  

Amazon’s cloud done been down, off and on, for a couple of days now.  

ReadWriteWeb’s Alex Williams is keepin’ ’em honest, and updated, which apparently Amazon hasn’t been too busy doing.

ComputerWorld describes it as an outright black eye for the “dominant player in the cloud market,” with sites like Quora, Reddit, and foursquare “left staggering or totally knocked out” because of server problems in the Amazon datacenter.

By way of example, Reddit’s site explained it was in “emergency read-only mode” because of Amazon’s degradation.

In the ComputerWorld piece, analyst Rob Enderle suggests the biggest impact from the outage may be to the cloud itself.

“What will take a hit is the image of this technology as being on you can depend on, and that image was critically damaged today,” Enderle is quoted as saying.

On the other hand, it should be noted there are lots of flavors of cloud computing, and a fully public, rentable cloud service like Amazon EC2 doesn’t even pretend to offer 100% availability.  In fact, it quite clearly states in its public FAQs that “The Amazon EC2 SLA guarantees 99.95% availability of the service within a Region over a trailing 365 day period.”

In any case, I would recommend carefully reading all the FAQs before fully banking one’s business in the public cloud.

And I would also recommend reading more widely about cloud implementations around the globe.

One size doesn’t fit all (there’s the public cloud, the private cloud, the hybrid, etc.), and the IBM Academy of Technology conducted a survey released this past October entitled “Cloud computing insights from 110 implementation projects” which explains pros and cons, the good and the bad, as well as related considerations to be thoughtfully considered before embarking upon cloud deployments.

Maybe read a little of that before the media or the blogosphere scare you back into your private data center bunker with your AR-15 and 10-day survival kits!

Written by turbotodd

April 22, 2011 at 8:29 pm